CN103929391A - Frequency calibration method and device - Google Patents

Abstract

The invention relates to the communication technology and discloses a frequency calibration method and device. A terminal sends a PSK modulation signal, after the PSK modulation signal is received, a section of data long enough behind the position of the rising edge is intercepted, a group of data are determined for rough frequency offset estimation and rough frequency offset compensation, wherein variance of the amplitude of the data is the smallest, then the precise time slot starting position is determined, data of a time slot are intercepted again, and a fine frequency offset deviant is determined according to the data, so that the accurate frequency offset is obtained, and accuracy of frequency calibration can be improved by conducting frequency calibration according to the frequency offset. Meanwhile, due to the fact that rough frequency offset estimation and rough frequency offset compensation are first conducted, then the precise time slot starting position is determined, and the data are intercepted for fine frequency offset compensation, efficiency for determining the frequency offset deviant is improved, and efficiency of frequency calibration is further improved.

Description

A kind of transmitting frequency calibration method and device

Technical field

The present invention relates to the communication technology, relate in particular to a kind of transmitting frequency calibration method and device.

Background technology

Terminal all needs to carry out RF consistency test before networking, terminal is carried out to fast frequency calibration and become one of prerequisite function of testing equipment, feature at the frequency error of calibration phase terminal is that frequency error range is large, can reach tens kHz, thereby testing equipment has been proposed to higher requirement.

Along with the fast development of mobile communication system, support the terminal of multiple types also to obtain developing rapidly, manufacturer terminal has been developed the Related product of supporting the multiple types such as GSM, WCDMA, TD-SCDMA, LTE at present.Before these end products network, need to carry out uniformity test to it, and RF consistency test is wherein the most basic test wherein, is the basis of carrying out other test.

What at present the terminal of these different systems is carried out to RF consistency test is generally the testing equipment combining, and is called as comprehensive test instrument.Comprehensive test instrument generally carries out, the test about measuring item, also needing to possess quickly calibrated function except possessing according to the related protocol of RF consistency test, terminal is carried out to the calibration of frequency, power and gain.The calibration of frequency is commonly referred to as to AFC (Automatic Frequency Control, automatic frequency control) calibration, often needs frequency error in a big way to estimate fast and accurately.

AFC calibration generally can be divided into two kinds of patterns: one is CW wave mode, and one is modulation signal pattern.Under CW wave mode, what comprehensive test instrument received is the known tone signal of carrier frequency, like this because reference signal is known, and is single, do not need to carry out Timing Synchronization to have the process of the recovery reference signal of high request, thus fairly simple to the estimation of frequency deviation.And under modulation signal pattern, even if the signal of pilot tone or training sequence is known, but because its feature is not single, but a string sampled point with modulation intelligence, therefore frequency error is estimated very responsive to Timing Synchronization error, need to carry out accurate Timing Synchronization, can match as much as possible to make reference signal and to receive signal, estimate thereby can carry out enough frequency error accurately.

Essence under modulation signal pattern is synchronously generally by pilot tone or training sequence being correlated with and correlation search obtains, but in the time that frequency error is larger, can make conversely again essence synchronously inaccurate, so just formed contradiction between the two.In order to solve this contradiction, general way is by large frequency deviation step-length is set, the modulation signal receiving to be compensated, and re-uses pilot tone or training sequence carries out the calculating of correlation and the judgement of correlation peak validity after each compensation.Do like this and carry out on the one hand the thresholding of peak value validity while judge and be difficult to setting, on the other hand, owing to carrying out large frequency deviation search according to a fixed step size, also can increase processing delay and have the factor of indeterminable delay

Visible, frequency deviation algorithm for estimating when comprehensive test instrument AFC under existing modulation signal pattern calibrates, there is the contradiction of the large and essence of frequency offset estimation range between synchronous, in order to solve that this contradiction is generally carried out large frequency deviation search by a fixed step size until the correlation peak of essence when synchronous meet certain thresholding and be judged as effectively, in fact the thresholding of peak value validity is difficult to set so on the one hand, and large frequency deviation search also can increase processing delay and have the factor of indeterminable delay.

Summary of the invention

The embodiment of the present invention provides a kind of transmitting frequency calibration method and device, to improve the accuracy of frequency calibration and the efficiency of frequency calibration.

A kind of transmitting frequency calibration method, comprising:

The psk modulation signal that receiving terminal sends, and the rising edge position of definite described signal;

According to rising edge position data intercept, and according to over-sampling rate, the data that intercept are divided into groups;

Determine the grouping of the variance minimum of amplitude, determine thick frequency deviation deviant according to the data in this grouping, and carry out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to described amplitude;

Determine time slot original position according to pilot tone or training sequence, intercept the data of a time slot and determine thin frequency deviation deviant according to these data;

According to described thick frequency deviation deviant and described thin frequency deviation deviant and, carry out frequency calibration.

A kind of frequency calibration device, comprising:

Rising edge determining unit, the psk modulation signal sending for receiving terminal, and the rising edge position of definite described signal;

Grouped element, for according to rising edge position data intercept, and divides into groups to the data that intercept according to over-sampling rate;

Thick frequency deviation estimating unit, for determining the grouping of variance minimum of amplitude, determines thick frequency deviation deviant according to the data in this grouping, and carries out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to described amplitude;

Thin frequency deviation estimating unit, for determining time slot original position according to pilot tone or training sequence, intercepts the data of a time slot and determines thin frequency deviation deviant according to these data;

Frequency calibration unit, for according to described thick frequency deviation deviant and described thin frequency deviation deviant and, carry out frequency calibration.

The embodiment of the present invention provides a kind of transmitting frequency calibration method and device, make terminal send psk modulation signal, receive after psk modulation signal, intercept sufficiently long one piece of data behind rising edge position, one group of data determining the variance minimum of amplitude carry out that thick frequency deviation is estimated and thick compensate of frequency deviation, determine again time slot original position more accurately, again intercept the data of a time slot, determine thin frequency deviation deviant according to these data, thereby obtain frequency deviation skew more accurately, carry out frequency calibration according to this frequency deviation skew and can improve the accuracy of frequency calibration, simultaneously, owing to first carrying out, thick frequency deviation is estimated and thick compensate of frequency deviation, determine again time slot original position more accurately, data intercept carries out thin compensate of frequency deviation, so improved the efficiency of definite frequency deviation deviant, and then improve the efficiency of frequency calibration.

Brief description of the drawings

The transmitting frequency calibration method flow chart that Fig. 1 provides for the embodiment of the present invention;

The more concrete transmitting frequency calibration method flow chart that Fig. 2 provides for the embodiment of the present invention;

The transmitting frequency calibration method flow chart of the corresponding embodiment mono-that Fig. 3 provides for the embodiment of the present invention;

The transmitting frequency calibration method flow chart of the corresponding embodiment bis-that Fig. 4 provides for the embodiment of the present invention;

The frequency calibration apparatus structure schematic diagram that Fig. 5 provides for the embodiment of the present invention.

Embodiment

The embodiment of the present invention provides a kind of transmitting frequency calibration method and device, make terminal send psk modulation signal, receive after psk modulation signal, intercept sufficiently long one piece of data behind rising edge position, one group of data determining the variance minimum of amplitude carry out that thick frequency deviation is estimated and thick compensate of frequency deviation, determine again time slot original position more accurately, again intercept the data of a time slot, determine thin frequency deviation deviant according to these data, thereby obtain frequency deviation skew more accurately, carry out frequency calibration according to this frequency deviation skew and can improve the accuracy of frequency calibration, simultaneously, owing to first carrying out, thick frequency deviation is estimated and thick compensate of frequency deviation, determine again time slot original position more accurately, data intercept carries out thin compensate of frequency deviation, so improved the efficiency of definite frequency deviation deviant, and then improve the efficiency of frequency calibration.

As shown in Figure 1, the transmitting frequency calibration method that the embodiment of the present invention provides, comprising:

The psk modulation signal that step S101, receiving terminal send, and the rising edge position of definite signal;

Step S102, according to rising edge position data intercept, and according to over-sampling rate to intercept data divide into groups;

Step S103, determine the grouping of the variance minimum of amplitude, determine thick frequency deviation deviant according to the data in this grouping, and carry out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to amplitude;

Step S104, determine time slot original position according to pilot tone or training sequence, the data that intercept a time slot are also determined thin frequency deviation deviant according to these data;

Step S105, according to thick frequency deviation deviant and thin frequency deviation deviant and, carry out frequency calibration.

Visible, owing to first carrying out, thick frequency deviation is estimated and thick compensate of frequency deviation, then after definite time slot original position more accurately, the data that intercept a time slot are carried out thin frequency deviation estimation, thereby obtain frequency deviation deviant more accurately, have improved accuracy and the efficiency of frequency calibration.

Concrete, as shown in Figure 2, the method comprises:

Step S201, host computer configurating terminal send the PSK of M unit (such as BPSK, QPSK, 8PSK modulation etc.) modulation signal, the object of configuration is the modulation signal that can obtain the standard M PSK of unit planisphere like this, to facilitate removal modulation intelligence, thereby carry out frequency deviation estimation;

Step S202, comprehensive test instrument carry out power searching, determine the rising edge position that receives signal;

Step S203, according to rising edge position, by enough large surplus is set, comprehensive test instrument intercepts sufficiently long definite active data (determining the data after real rising edge), and data intercept is carried out to packet according to over-sampling rate;

Step S204, packet is asked respectively to the variance of its amplitude, variance minimum be grouped into optimum sampling grouping;

Step S205, ask thick frequency deviation: first to optimum sampling grouping ask M power (M is modulation index: for BPSK, M=2; For QPSK, M=4, for 8PSK, M=8; By that analogy) operation is to eliminate the modulation intelligence of M unit psk modulation signal; Afterwards to the M power obtaining (M is modulation index: for BPSK, M=2; For QPSK, M=4, for 8PSK, M=8; Data padding by that analogy), carry out FFT conversion, again the data after FFT conversion are carried out to FFT shifting function to make zero-frequency in center, wherein need to select counting of suitable FFT conversion to make FFT resolution at least within the scope of following thin frequency deviation estimation; Then ask the peak of the absolute value of the data after FFT shifting function; Finally obtain thick frequency deviation deviant according to the value of the difference of peak and center and FFT resolution and M;

Step S206, optimum sampling grouping is carried out to thick compensate of frequency deviation;

Step S207, obtain local pilot data according to test configurations, use the data after local pilot data and thick compensate of frequency deviation to carry out correlation value calculation, and find the peak of the absolute value of correlation;

The peak that step S208, basis obtain intercepts the data of a time slot accurately, and time slot data are detected to the symbol data that obtains a time slot with demodulation;

Step S209, to symbol data ask M power (M is modulation index: for BPSK, M=2; For QPSK, M=4, for 8PSK, M=8; By that analogy) thus remove modulation intelligence, then to M power (M is modulation index: for BPSK, M=2; For QPSK, M=4, for 8PSK, M=8; Data by that analogy) are used approximate maximal possibility estimation and divided by M, thereby obtain thin frequency deviation deviant;

Step S210, thick frequency deviation deviant and thin frequency deviation deviant are added, obtain final frequency offset estimation result.

Concrete, in WCDMA system, can be configured terminal, make terminal only send DPCCH signal, do not send DPDCH signal, thereby obtain the qpsk modulation signal of standard, be convenient to remove modulation intelligence and carry out frequency deviation estimation.

In step S101, receive the psk modulation signal that terminal sends, and the rising edge position of definite signal, can be by determining the performance number of psk modulation signal sampling point sliding window, then determine that according to the performance number of the sampling point sliding window of each sampled point the rising edge position of signal realizes.

In step S102, according to rising edge position data intercept, and according to over-sampling rate, the data that intercept are divided into groups, specifically comprise:

Determine data intercept be r '=[r ' (0), r ' (1) ..., r ' (N '-1)], N '=OSR*N _fOE, wherein r ' (i)=r (i+I _start+ N _Δ* OSR), OSR is over-sampling rate, N _Δfor the surplus setting in advance, N _fOE≤ N _cfor the symbolic number of estimating in order to frequency deviation, N _cbe the number of chips of a time slot, wherein, for WCDMA, N _cbe 2560, for TD-SCDMA, N _cbe 864.

To the front N of the data that intercept _var* OSR data are divided into groups, and obtain OSR group r ₀', r ' ₁..., r ' _oSR-1, equal 1 sampling data in every group, wherein, N _varfor the number of chips of each grouping, and N _var≤ N _fOE.

In step S103, determine the grouping of the variance minimum of amplitude, determine thick frequency deviation deviant according to the data in this grouping, specifically comprise:

The operation of M power is carried out in the grouping of the variance minimum to amplitude, and wherein, M is modulation index;

To the data padding after the M power operation obtaining, carry out FFT conversion, wherein, counting in setting range of FFT conversion, makes FFT resolution in the scope of thin frequency deviation estimation;

Data after FFT conversion are carried out to FFT shifting function, make zero-frequency in center;

Determine the peak of the absolute value of the data after FFT shifting function;

Obtain thick frequency deviation deviant according to the value of the difference of peak and center and FFT resolution and M.

Concrete, the grouping of the variance minimum to amplitude is carried out after the operation of M power, removes modulation intelligence, obtains r ' _pow4=[r ' _pow4(0), r ' _pow4(1) ..., r ' _pow4(N _fOE-1)], in the time that this signal is QPSK signal, M=4, r ' _pow4(i)=(r ' _tFFT(i)) ⁴=(r ' _fFT(i)) ²(r ' _fFTT(i)) ²;

To r ' _pow4zero padding obtains N _fFTpoint carries out the data of FFT: wherein N _fFTit is 2 integral number power;

Carry out FFT computing and obtain r _{fFT_OUT}=FFT (r _{fFT_IN});

Carry out FFT shifting function and obtain r _{fFT_Shift}=[r _{fFT_Shift}(0), r _{fFT_Shift}(1) ..., r _{fFT_Shift}(N _fFT-1)], wherein,

$r_{\mathrm{FFT}\_\mathrm{Shift}}\left(i\right)=\left\{\begin{array}{c}{r}_{\mathrm{FFT}\_\mathrm{OUT}}(i+\frac{N_{\mathrm{FFT}}}{2}),0\≤i\≤\frac{N_{\mathrm{FFT}}}{2}\\ r_{\mathrm{FFT}\_\mathrm{OUT}}(i-\frac{N_{\mathrm{FFT}}}{2}),\frac{N_{\mathrm{FFT}}}{2}\≤i\≤N_{\mathrm{FFT}}\end{array}\right.;$

Determine r _{fFT_Shift}the position of absolute value maximum $A_{\mathrm{index}}=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{FFT}}}{\max }\left(\right|r_{\mathrm{FFT}\_\mathrm{Shift}}\left(i\right)\left|\right);$

Determine that thick frequency offset estimation result is wherein for the frequency resolution of FFT conversion; f _cbe 1 sampling speed.

In step S103, carry out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to amplitude, specifically comprise:

Determine that the data after thick compensate of frequency deviation are r _comp=[r _comp(0), r _comp(1) ..., r _comp(N _comp-1)], wherein, $r_{\mathrm{comp}}\left(i\right)=r(I_{\mathrm{start}}+I_{\mathrm{opt}}+i*\mathrm{OSR})e^{-j2\mathrm{\πi}{f}_{\mathrm{el}}{T}_c},i=\mathrm{0,1},...,N_{\mathrm{comp}}-1;$ T _cfor chip period, N _comp=N _c+ N _sync, N _cbe the number of chips (being 2560 for WCDMA, is 864 for TD-SCDMA) of 1 time slot, N _syncfor carrying out the data window length of synchronous search, f _e1for thick frequency deviation deviant, I _optfor the sampling deviation of this grouped data.It should be noted that the rising edge position I that hypothesis is estimated here _startdetermine less than real rising edge position, to make within real rising edge position is included in sync search window; Otherwise, need to increase a surplus, to make within real rising edge position is included in sync search window.

In WCDMA system, need to determine time slot original position according to pilot tone, and in TD-SCDMA system, need to determine time slot original position according to training sequence.

Concrete, in the time that signal is WCDMA signal, in step S104, determine time slot original position according to pilot tone, specifically comprise:

According to the pilot frequency format of configuration, obtain original pilots bit:

d _pilot＝[d _pilot(0)，d _pilot(1)，…，d _pilot(N _pilot-1)]；

By d _pilotcarry out BPSK modulation and obtain modulation symbol d ' _pilot=[d ' _pilot(0), d ' _pilot(1) ..., d ' _pilot(N _pilot-1)], d ' wherein _pilot(i)=-2d _pilot(i)+1;

To d ' _pilotcarry out spread spectrum: $d_{\mathrm{spr},\mathrm{pilot}}=d_{\mathrm{pilot}}^{\′}\⊗c_{\mathrm{ch},\mathrm{256,0}}=[d_{\mathrm{pilot}}^{\′}\left(0\right)*c_{\mathrm{ch},\mathrm{256,0}},d_{\mathrm{pilot}}^{\′}\left(1\right)*c_{\mathrm{ch},\mathrm{256,0}},...,d_{\mathrm{pilot}}^{\′}(N_{\mathrm{pilot}}-1)*c_{\mathrm{ch},\mathrm{256,0}}],$ Wherein, represent direct product, c _{ch, 256,0}for spreading code.

By the scrambler s of this time slot ₀=[s ₀(0), s ₀(1) ..., s ₀(2559)] carry out scrambling and obtain reference pilot signal: d _{scr, pilot}=[d _{scr, pilot}(0), d _{scr, pilot}(1) ..., d _{scr, pilot}(256*N _pilot-1)], d wherein _{scr, pilot}(i)=(j*d _{spr, pilot}(i)) * s ₀(i);

Determine correlation R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)], wherein,

$R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{N_{c,\mathrm{pilot}}-1}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j)d_{\mathrm{scr},\mathrm{pilot}}*\left(j\right);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is: r (I _start+ I _opt+ I _max* OSR);

In the time that signal is TD-SCDMA signal, in step S104, determine time slot original position according to training sequence (intermediate code), specifically comprise:

Obtain with reference to intermediate code symbol with intermediate code deviant according to the basic midamble code of configuration number:

d _mid＝[d _mid(0)，d _mid(1)，…，d _mid(N _mid-1)]，

Wherein, N _mid=144 is intermediate code symbolic number, d _mid(i) 1 ,-1, j, value in-j};

Determine correlation: R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)],

Wherein, $R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{127}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j+352+16)d_{\mathrm{mid}}*(j+16);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is: r (I _start+ I _opt+ I _max* OSR).

In the time that signal is WCDMA signal, in step S104, determine thin frequency deviation deviant according to these data, specifically comprise:

These data are carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence.

Concrete, first determine the data r from thick compensate of frequency deviation _compthe data of middle intercepting are: r _{slot, 0}=[r _{slot, 0}(0), r _{slot, 0}(1) .., r _{slot, 0}(N _c-1)], r wherein _{slot, 0}(i)=r _comp(i+I _max);

These data are carried out to descrambling and obtain d _{des, dpcch}=[d _{des, dpcch}(0), d _{des, dpcch}(1) ..., d _{des, dpcch}(N _c-1)], d wherein _{des, dpcch}(i)=r _{slot, 0}(i) * s ₀* (i)

Data after descrambling are carried out to despreading and obtain d _soft=[d _soft(0), d _soft(1) ..., d _soft(9)],

Wherein, $d_{\mathrm{soft}}\left(i\right)=\frac{1}{256}\underset{j=0}{\overset{255}{\mathrm{\Σ}}}{d}_{\mathrm{des},\mathrm{dpcch}}(i*256+j)*c_{\mathrm{ch},\mathrm{256,0}}\left(j\right)=\frac{1}{256}\underset{j=0}{\overset{255}{\mathrm{\Σ}}}{d}_{\mathrm{des},\mathrm{dpcch}}(i*256+j);$

Remove the modulation intelligence in the data after despreading, obtain: d _sqr=[d _sqr(0), d _sqr(1) ..., d _sqr(N _sym-1)], in the time that this signal is bpsk signal, d _sqr(i)=(d _soft(i)) ², N _symit is the DPCCH number of symbols of 1 time slot;

Determine d _sqrthe Delay autocorrelation of middle element: $R=\frac{1}{N_{\mathrm{sym}}-1}\underset{i=1}{\overset{N_{\mathrm{sym}}-1}{\mathrm{\Σ}}}{d}_{\mathrm{sqr}}\left(i\right){d*}_{\mathrm{sqr}}(i-1);$

Determine that thin frequency deviation deviant is: wherein T _sfor the symbol period of DPCCH channel; In the time that signal is TD-SCDMA signal, in step S104, determine thin frequency deviation deviant according to these data, specifically comprise:

Get the result after the thick compensate of frequency deviation of a time slot;

Intercept the data block of Midamble (centre) code both sides;

The data block intercepting is carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence.

Concrete, according to the smart sync bit obtaining, the data r from thick compensate of frequency deviation _compthe data of a time slot of middle intercepting: r _slot=[r _slot(0), r _slot(1) ..., r _slot(N _c-1)], r wherein _slot(i)=r _comp(i+I _max);

The data block that intercepts intermediate code both sides, first data block is expressed as d ₁=[d ₁(0), d ₁(1) ..., d ₁(351)], second data block is expressed as d ₂=[d ₂(0), d ₂(1) ..., d ₂(351)], d wherein ₁(i)=r _{slot, 0}(i), d ₂(i)=r _{slot, 0}(i+352+144);

Descrambling:

TD-SCDMA uses the scrambler that length is 16, determines plural scrambler sequence according to the scrambling code number of configuration, is designated as: s=[s (0), and s (1) ..., s (15)];

To d ₁descrambling obtains: d _des1=[d _des1(0), d _des1(1) ..., d _des1(351)], d wherein _des1(i)=d ₁(i) * s ^*(mod (i, 16)); To d ₂descrambling obtains: d _des2=[d _des2(0), d _des2(1) ..., d _des2(351)], d wherein _des2(i)=d ₂(i) * s ^*(mod) (i, 16)); Despreading:

The spreading factor that establishing is put is SF _dpch, the symbolic number of data block 1 and data block 2 is: N _sym=352/SF _dpch;

According to configuration, the spreading code (the spreading code coefficient containing according to protocol package) that obtains using is: c _dpch=[c _dpch(0), c _dpch(1) ..., c _dpch(SF _dpch-1)].

To d _desldespreading obtains: d _{1, soft}=[d _{1, soft}(0), d _{1, soft}(1) ..., d _{1, soft}(N _sym-1)],

Wherein $d_{1,\mathrm{soft}}\left(i\right)=\underset{j=0}{\overset{{\mathrm{SF}}_{\mathrm{dpch}}-1}{\mathrm{\Σ}}}{d}_{\mathrm{des}1}(i*{\mathrm{SF}}_{\mathrm{dpch}}+j)*c_{\mathrm{dpch}}^{*}\left(\mathrm{mod}(j,{\mathrm{SF}}_{\mathrm{dpch}})\right);$

To d _des2despreading obtains: d _{2, soft}=[d _{2, soft}(0), d _{2, soft}(1) ..., d _{2, soft}(N _sym-1)],

Wherein $d_{2,\mathrm{soft}}\left(i\right)=\underset{j=0}{\overset{{\mathrm{SF}}_{\mathrm{dpch}}-1}{\mathrm{\Σ}}}{d}_{\mathrm{des}2}(i*{\mathrm{SF}}_{\mathrm{dpch}}+j)*c_{\mathrm{dpch}}^{*}\left(\mathrm{mod}(j,{\mathrm{SF}}_{\mathrm{dpch}})\right);$

Remove modulation intelligence:

Different from WCDMA, the DPCH data d of descrambling, despreading _{1, soft}and d _{2, soft}be still qpsk modulation signal, can adopt and ask the mode of 4 powers to remove modulation intelligence, obtain:

d _1，pow4＝[d _1，pow4(0)，d _1，pow4(1)，...，d _1，pow4(N _sym-1)]，

d _2，pow4＝[d _2，pow4(0)，d _2，pow4(1)，...，d _2，pow4(N _sym-1)]。

Wherein $d_{k,\mathrm{pow}4}\left(i\right)=d_{k,\mathrm{soft}}^4\left(i\right),i=\mathrm{0,1},...,N_{\mathrm{sym}}-1,k=\mathrm{1,2};$

Calculate the correlation of two data blocks: $R=\underset{i=1}{\overset{N_{\mathrm{sym}}}{\mathrm{\Σ}}}{d}_{1,\mathrm{sqr}}\left(i\right){d*}_{2,\mathrm{sqr}}\left(i\right);$

Calculate residual frequency deviation: $f_{e2}=\frac{1}{2\mathrm{\π}{T}_s(N_{\mathrm{sym}}+N_{\mathrm{mid}})}\arg \left(R\right)/4,$ Wherein $T_s=\mathrm{SF}*T_c=\frac{\mathrm{SF}}{1.28\×{10}^6}$ Second, be the symbol period of DPCH channel, N _midfor the number under character rate corresponding to the time span of midamble code, be 144/SF.

Below, respectively the transmitting frequency calibration method in WCDMA system and in TD-SCDMA system is elaborated by specific embodiment:

Embodiment mono-,

This embodiment is in WCDMA system, frequency calibration when AFC calibrates under modulation signal pattern.

As shown in Figure 3, this transmitting frequency calibration method comprises:

Step S301, host computer configuration β _c=15 and β _d=0, thus make terminal only send DPCCH signal and do not send DPDCH signal, and the object of configuration is the modulation signal that can obtain standard QPSK planisphere like this, to facilitate removal modulation intelligence, thereby carries out frequency deviation estimation.

Step S302, rising edge detect:

Suppose that comprehensive test instrument receives the data r=[r (0) of a radio frames, r (1) ..., r (N-1)], N=OSR*N _c* N _slot, wherein OSR is over-sampling rate, N _c=2560 is the number of chips of a time slot of DPCCH channel, N _slot=15 number of time slot for input data;

Calculate the performance number that receives signal sampling point sliding window: $P_{\mathrm{win}}\left(i\right)=\underset{n=i}{\overset{i+L_{\mathrm{win}}-1}{\mathrm{\Σ}}}{\left|r\left(n\right)\right|}^2,i=\mathrm{0,1},...,N-L_{\mathrm{win}};$ Wherein L _winthe sampling point number comprising for power calculation window length.

Search rising edge:

$\begin{array}{ccc}\mathrm{for}& & i=0:N-L_{\mathrm{win}}\\ & \mathrm{if}& \frac{P_{\mathrm{win}}(i+L_{\mathrm{win}})}{P_{\mathrm{win}}\left(i\right)}>P_{\lim }\\ & & I_{\mathrm{start}}=i+2L_{\mathrm{win}}\\ & \mathrm{end}& \\ & \mathrm{break};& \\ \mathrm{end}& & \end{array};$

Wherein P _limfor rising edge difference power threshold value.

If known received signal power value, the condition of rising edge judgement also can adopt power absolute value.

Step S303, intercepting valid data, and divide into groups:

Take out in order to carry out the data of thick frequency deviation estimation:

r′＝[r′(0)，r′(1)，…，r′(N′-1)]，N′＝OSR*SF _dpcch*N _FOE，

Wherein r ' (i)=r (i+I _start+ N _Δ* OSR), N _Δthe surplus arranging in order to guarantee to get the data after the position that authentic data starts, N _fOE≤ N _cfor the symbolic number of estimating in order to frequency deviation, N _cit is the number of chips (being 2560 for WCDMA, is 864 for TD-SCDMA) of a time slot.

Packet:

To the front N of r ' _var* OSR data are divided into groups, and obtain OSR and organize 1 sampling data:

R ' ₀, r ' ₁..., r ' _oSR-1, wherein N _var≤ SF _dpcch* N _foe1for the number of chips of each grouping.

J sampling point value of i group data is:

r′ _i(j)＝r′(j*OSR+i)，i＝0，1，...，OSR-1，j＝0，1，...，N _var-1。

Step S304, determine optimum sampling grouping:

To element in each group of data ask respectively amplitude square, obtain: r ' _{s, 0}, r ' _{s, 1}..., r ' _{s, OSR-1},

Wherein r ' _{s, i}(j)=| r ' _i(j) | ², i=0,1 ..., OSR-1, j=0,1 ..., N _var-1;

Amplitude to each group of data square, average respectively, obtain:

Wherein $\stackrel{\‾}{r_{s,i}^{\′}}=\frac{1}{N_{\mathrm{var}}}\underset{j=0}{\overset{N_{\mathrm{var}}-1}{\mathrm{\Σ}}}{r}_{s,i}^{\′}\left(j\right),i=\mathrm{0,1},...,\mathrm{OSR}-1.$

Ask variance:

The variance of asking each mould side's vector, obtains σ ₀, σ ₁..., σ _oSR-1,

Wherein ${\mathrm{\σ}}_i=\frac{1}{N_{\mathrm{var}}}\underset{j=0}{\overset{N_{\mathrm{var}}-1}{\mathrm{\Σ}}}{(r_{s,i}^{\′}\left(j\right)-\stackrel{\‾}{r_{s,i}^{\′}})}^2,i=\mathrm{0,1},...,\mathrm{OSR}-1.$

Ask optimum sampling position:

The sampling deviation minimum of the minimum Yi of variance road grouped data:

Thick frequency deviation estimation is carried out in step S305, use FFT computing:

Intercept in order to carry out the sampling point of FFT computing: r ' _fFT=[r ' _fFT(0), r ' _fFT(1) ..., r ' _fFT(N _fOE-1)], r ' wherein _fFT(i)=r (i*OSR+I _start+ N _Δ* OSR+I _opt), i=0,1 ..., N _fOE-1, N _fOEfor carrying out effective number of samples of FFT computing;

Because r ' _fFTfor qpsk modulation signal, it is asked to 4 power operations, can eliminate the phase hit that modulation brings, obtain:

r′ _pow4＝[r′ _pow4(0)，r′ _pow4(1)，...，r′ _pow4(N _FOE-1)]，

Wherein r ' _pow4(i)=(r ' _fFTT(i)) ⁴=(r ' _fFT(i)) ²(r ' _fFT(i)) ².

To r ' _pow4zero padding obtains N _fFTpoint carries out the data of FFT:

Wherein N _fFTit is 2 integral number power.

Carry out FFT computing, and FFT shifting function:

FFT computing obtains: r _{fFT_OUT}=FFT (r _{fFT_IN});

FFT displacement obtains: r _{fFT_Shift}=[r _{fFT_Shifi}(0), r _{fFT_Shift}(1) ..., r _{fFT_Shift}(N _fFT-1)];

Wherein $r_{\mathrm{FFT}\_\mathrm{Shift}}\left(i\right)=\left\{\begin{array}{c}{r}_{\mathrm{FFT}\_\mathrm{OUT}}(i+\frac{N_{\mathrm{FFT}}}{2}),0\≤i\≤\frac{N_{\mathrm{FFT}}}{2}\\ r_{\mathrm{FFT}\_\mathrm{OUT}}(i-\frac{N_{\mathrm{FFT}}}{2}),\frac{N_{\mathrm{FFT}}}{2}\≤i\≤N_{\mathrm{FFT}}\end{array}\right..$

Look for the position of absolute value maximum: $A_{\mathrm{index}}=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{FFT}}}{\max }\left(\right|r_{\mathrm{FFT}\_\mathrm{Shift}}\left(i\right)\left|\right).$

Thick frequency deviation is estimated: $f_{e1}=(A_{\mathrm{index}}-\frac{N_{\mathrm{FFT}}}{2})*\mathrm{\Δf}/4,$

Wherein for the frequency resolution of FFT conversion; f _cbe 1 sampling speed (equaling spreading rate), value is 3.84 × 10 ⁶hz; Eliminate 4 powers divided by 4 and frequency deviation has been expanded to the impact of 4 times.

Step S306, thick compensate of frequency deviation, obtain: r _comp=r _comp(0), r _comp(1) ..., r _comp(M _comp-1)],

Wherein $r_{\mathrm{comp}}\left(i\right)=r(I_{\mathrm{start}}+I_{\mathrm{opt}}+i*\mathrm{OSR})e^{-j2\mathrm{\πi}{f}_{\mathrm{el}}{T}_c},i=\mathrm{0,1},...,N_{\mathrm{comp}}-1;$ T _cfor WCDMA chip period; N _comp=N _c+ N _sync, N _c=2560 is the number of chips of 1 time slot, N _syncfor carrying out the data window length (number of chips) of synchronous search;

Calculating and the carefully estimation of frequency deviation that data after thick compensate of frequency deviation are synchronously asked correlation in order to carry out essence.Step S307, smart synchronous:

After reference pilot signal generates, according to the pilot frequency format of configuration, obtain original pilots bit:

d _pilot＝[ _dpilot(0)，d _pilot(1)，…，d _pilot(N _pilot-1)]；

By d _pilotcarry out BPSK modulation and obtain modulation symbol d ' _pilot=[d ' _pilot(0), d ' _pilot(1) ..., d ' _pilot(N _poilt-1)], d ' wherein _pilot(i)=-2d _pilot(i)+1, value is+1 or-1;

By d ' _pilotuse spreading code c _{ch, 256,0}spread spectrum, wherein c _{ch, 256,0}the spreading code that the DPCCH channel specifying for agreement uses, value is complete 1 256 n dimensional vector ns: $d_{\mathrm{spr},\mathrm{pilot}}=d_{\mathrm{pilot}}^{\′}\⊗c_{\mathrm{ch},\mathrm{256,0}}=[d_{\mathrm{pilot}}^{\′}\left(0\right)*c_{\mathrm{ch},\mathrm{256,0}},d_{\mathrm{pilot}}^{\′}\left(1\right)*c_{\mathrm{ch},\mathrm{256,0}},...,d_{\mathrm{pilot}}^{\′}(N_{\mathrm{pilot}}-1)*c_{\mathrm{ch},\mathrm{256,0}}],$ Wherein represent direct product;

Scrambling obtains reference pilot signal d _{scr, pilot};

The scrambler that obtains this time slot according to the scrambler ID of host computer configuration is: s ₀=[s ₀(0), s ₀(1) ..., s ₀(2559)].

Scrambling obtains: d _{scr, pilot}=[d _{scr, pilot}(0), d _{scr, pilot}(1) ..., d _{scr, pilot}(256*N _pilot-1)], d wherein _{scr, pilot}(i)=(j*d _{spr, pilot}(i)) * s ₀(i),

Correlation value calculation, obtains R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)], wherein

$R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{N_{c,\mathrm{pilot}}-1}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j)d_{\mathrm{scr},\mathrm{pilot}}*\left(j\right);$

The peak of correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$ The accurate original position of time slot is: sampling point r (I _start+ I _opt+ I _max* OSR).

Step S308, thin frequency deviation are estimated:

According to the smart sync bit obtaining, the data r from thick compensate of frequency deviation _compthe data of a time slot of middle intercepting: r _{slot, 0}=[r _{slot, 0}(0), r _{slot, 0}(1) ..., r _{slot, 0}(N _c-1)], r wherein _{slot, 0}(i)=r _comp(i+I _max);

Descrambling obtains: d _{des, dpcch}=[d _{des, dpcch}(0), d _{des, dpcch}(1) ..., d _{des, dpcch}(N _c-1)], d wherein _{des, dpcch}(i)=r _{slot, 0}(i) * s ₀* (i)

Despreading obtains: d _soft=[d _soft(0), d _soft(1) ..., d _soft(9)],

Wherein, $d_{\mathrm{soft}}\left(i\right)=\frac{1}{256}\underset{j=0}{\overset{255}{\mathrm{\Σ}}}{d}_{\mathrm{des},\mathrm{dpcch}}(i*256+j)*c_{\mathrm{ch},\mathrm{256,0}}\left(j\right)=\frac{1}{256}\underset{j=0}{\overset{255}{\mathrm{\Σ}}}{d}_{\mathrm{des},\mathrm{dpcch}}(i*256+j),$

Because the d obtaining _softfor the DPCCH data after descrambling, despreading, it is the BPSK modulation signal that is positioned at the imaginary axis, can adopt the mode of asking square to remove modulation intelligence, obtains: d _sqr=[d _sqr(0), d _sqr(1) ..., d _sqr(N _sym-1)], d wherein _sqr(i)=(d _soft(i)) ², N _sym=10, be the DPCCH number of symbols of 1 time slot;

Calculate d _sqrthe Delay autocorrelation R of middle element: $R=\frac{1}{N_{\mathrm{sym}}-1}\underset{i=1}{\overset{N_{\mathrm{sym}}-1}{\mathrm{\Σ}}}{d}_{\mathrm{sqr}}\left(i\right){d*}_{\mathrm{sqr}}(i-1);$

Calculate residual frequency deviation: wherein second, be the symbol period of DPCCH channel.

Step S309, determine total frequency deviation: f _e=f _e1+ f _e2;

Step S310, the total frequency deviation of basis are carried out frequency calibration.

Embodiment bis-,

This embodiment is in TD-SCDMA system, and frequency deviation when AFC calibrates under 12.2kbps modulation signal pattern is estimated, and supposition is doing before frequency deviation estimation, has passed through Timing Synchronization, and timing synchronization procedure is identical with timing synchronization procedure in embodiment mono-.

As shown in Figure 4, this transmitting frequency calibration method comprises:

Step S401, host computer configurating terminal send the DPCH signal of QPSK modulation at the ascending time slot distributing, in addition, host computer also needs the scrambling code number that configurating terminal sends, basic Midamble code (intermediate code) number, scrambling code number (equaling basic midamble code number according to agreement scrambling code number), intermediate code skew and spreading factor and spreading code.

Step S402, rising edge detect:

Suppose that comprehensive test instrument receives the data of a subframe lengths:

r＝[r(0)，r(1)，…，r(N-1)]，N＝OSR*N _subframe，

Wherein, OSR is over-sampling rate, N _subframe=6400 is the number of chips of a subframe of TD-SCDMA;

Calculate the performance number that receives signal sampling point sliding window: $P_{\mathrm{win}}\left(i\right)=\underset{n=i}{\overset{i+L_{\mathrm{win}}-1}{\mathrm{\Σ}}}{\left|r\left(n\right)\right|}^2,i=\mathrm{0,1},...,N-L_{\mathrm{win}};$ Wherein L _winthe sampling point number comprising for power calculation window length;

Search rising edge:

$\begin{array}{ccc}\mathrm{for}& & i=0:N-L_{\mathrm{win}}\\ & \mathrm{if}& \frac{P_{\mathrm{win}}(i+L_{\mathrm{win}})}{P_{\mathrm{win}}\left(i\right)}>P_{\lim }\\ & & I_{\mathrm{start}}=i+2L_{\mathrm{win}}\\ & \mathrm{end}& \\ & \mathrm{break};& \\ \mathrm{end}& & \end{array};$

Wherein P _limfor rising edge difference power threshold value.

If known received signal power value, the condition of rising edge judgement also can adopt power absolute value.

Step S403, intercepting valid data, and divide into groups:

Take out in order to carry out the data of thick frequency deviation estimation:

r′＝[r′(0)，r′(1)，…，r′(N′-1)]，N′＝OSR*N _FOE。

Wherein r ' (i)=r (i+I _start+ N _Δ* OSR), N _Δthe surplus arranging in order to guarantee to get the data after the position that authentic data starts, N _fOE≤ N _cfor the number of chips of estimating in order to frequency deviation, N _c=848 is the number of chips (not comprising defendance compartment) of a time slot.

Packet:

To the front N of r ' _var* OSR data are divided into groups, and obtain OSR and organize 1 sampling data:

R ' ₀, r ' ₁..., r ' _oSR-1, wherein N _var≤ N _fOEfor the number of chips of each grouping.

J sampling point value of i group data is:

r′ _i(j)＝r′(j*OSR+i)，i＝0，1，...，OSR-1，j＝0，1，...，N _var-1。

Step S404, determine optimum sampling grouping:

To element in each group of data ask respectively amplitude square, obtain: r ' _{s, 0}, r ' _{s, 1}..., r ' _{s, OSR-1},

Wherein r ' _{s, i}(j)=| r ' _i(j) | ², i=0,1 ..., OSR-1, j=0,1 ..., N _var-1.

Amplitude to each group of data square, average respectively, obtain:

Wherein $\stackrel{\‾}{r_{s,i}^{\′}}=\frac{1}{N_{\mathrm{var}}}\underset{j=0}{\overset{N_{\mathrm{var}}-1}{\mathrm{\Σ}}}{r}_{s,i}^{\′}\left(j\right),i=\mathrm{0,1},...,\mathrm{OSR}-1.$

Ask variance:

The variance of asking each mould side's vector, obtains σ ₀, σ ₁..., σ _oSR-1,

Wherein ${\mathrm{\σ}}_i=\frac{1}{N_{\mathrm{var}}}\underset{j=0}{\overset{N_{\mathrm{var}}-1}{\mathrm{\Σ}}}{(r_{s,i}^{\′}\left(j\right)-\stackrel{\‾}{r_{s,i}^{\′}})}^2,i=\mathrm{0,1},...,\mathrm{OSR}-1.$

Ask optimum sampling position:

The sampling deviation minimum of the minimum Yi of variance road grouped data:

Thick frequency deviation estimation is carried out in step S405, use FFT computing:

Intercept in order to carry out the sampling point of FFT computing: r ' _fFT=[r ' _fFT(0), r ' _fFT(1) ..., r ' _fFT(N _fOE-1)], r ' wherein _fFT(i)=r (i*OSR+I _start+ N _Δ* OSR+I _opt), i=0,1 ..., N _fOE-1, N _fOEfor carrying out effective number of samples of FFT computing;

Because r _fFTfor qpsk modulation signal, it is asked to 4 power operations, can eliminate the phase hit that modulation brings, obtain:

r′ _pow4＝[r′ _pow4(0)，r′ _pow4(1)，...，r′ _pow4(N _FOE-1)]，

Wherein r ' _pow4(i)=(r ' _fFT(i)) ⁴=(r ' _fFT(i)) ²(r ' _fFT(i)) ².

To r ' _pow4zero padding obtains N _fFTpoint carries out the data of FFT:

Wherein N _fFTit is 2 integral number power.

Carry out FFT computing, and FFT shifting function:

FFT computing obtains: r _{fFT_OUT}=FFT (r _{fFT_IN});

FFT displacement obtains: r _{fFT_Shift}=[r _{fFT_Shift}(0), r _{fFT_Shift}(1) ..., r _{fFT_Shift}(N _fFT-1)];

Wherein $r_{\mathrm{FFT}\_\mathrm{Shift}}\left(i\right)=\left\{\begin{array}{c}{r}_{\mathrm{FFT}\_\mathrm{OUT}}(i+\frac{N_{\mathrm{FFT}}}{2}),0\≤i\≤\frac{N_{\mathrm{FFT}}}{2}\\ r_{\mathrm{FFT}\_\mathrm{OUT}}(i-\frac{N_{\mathrm{FFT}}}{2}),\frac{N_{\mathrm{FFT}}}{2}\≤i\≤N_{\mathrm{FFT}}\end{array}\right.;$

Look for the position of absolute value maximum: $A_{\mathrm{index}}=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{FFT}}}{\max }\left(\right|r_{\mathrm{FFT}\_\mathrm{Shift}}\left(i\right)\left|\right).$

Thick frequency deviation is estimated: $f_{e1}=(A_{\mathrm{index}}-\frac{N_{\mathrm{FFT}}}{2})*\mathrm{\Δf}/4,$

Wherein for the frequency resolution of FFT conversion; f _cbe 1 sampling speed (equaling spreading rate), value is 1.28 × 10 ⁶hz; Eliminate 4 powers divided by 4 and frequency deviation has been expanded to the impact of 4 times.

Step S406, thick compensate of frequency deviation, obtain: r _comp=[r _comp(0), r _comp(1) ..., r _comp(N _comp-1)],

Wherein $r_{\mathrm{comp}}\left(i\right)=r(I_{\mathrm{start}}+I_{\mathrm{opt}}+i*\mathrm{OSR})e^{-j2\mathrm{\πi}{f}_{\mathrm{el}}{T}_c},i=\mathrm{0,1},...,N_{\mathrm{comp}}-1;$ T _cfor TD-SCDMA chip period; N _comp=N _c+ N _sync, N _c=848 is the number of chips (not comprising defendance compartment) of 1 time slot, N _syncfor carrying out the data window length (number of chips) of synchronous search.

Calculating and the carefully estimation of frequency deviation that data after thick compensate of frequency deviation are synchronously asked correlation in order to carry out essence.

Step S407, smart synchronous:

After local intermediate code (Midamble) signal generates, obtain intermediate code symbol according to the basic midamble code of configuration number with intermediate code deviant:

D _mid=[d _mid(0), d _mid(1) ..., d _mid(N _mid-1)], N wherein _mid=144 is intermediate code symbolic number, d _mid(i) 1 ,-1, j, value in-j}.

Correlation value calculation, obtains R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)], wherein

$R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{127}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j+352+16)d_{\mathrm{mid}}*(j+16);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of time slot is: sampling point r (I _start+ I _opt+ I _max* OSR).

Step S408, thin frequency deviation are estimated:

According to the smart sync bit obtaining, the data r from thick compensate of frequency deviation _compthe data of a time slot of middle intercepting: r _slot=[r _slot(0), r _slot(1) ..., r _slot(N _c-1)], r wherein _slot(i)=r _comp(i+I _max);

The data block that intercepts intermediate code both sides, first data block is expressed as d ₁=[d ₁(0), d ₁(1) ..., d ₁(351)], second data block is expressed as d ₂=[d ₂(0), d ₂(1) ..., d ₂(351)];

Wherein d ₁(i)=r _{slot, 0}(i), d ₂(i)=r _{slot, 0}(i+352+144);

TD-SCDMA uses the scrambler that length is 16, determines plural scrambler sequence according to the scrambling code number of configuration, is designated as: s=[s (0), and s (1) ..., s (15)];

To d ₁descrambling obtains: d _des1=[d _des1(0), d _des1(1) ..., d _des1(351)],

Wherein d _des1(i)=d ₁(i) * s ^*(mod (i, 16));

To d ₂descrambling obtains: d _des2=[d _des2(0), d _des2(1) ..., d _des2(351)],

Wherein d _des2(i)=d ₂(i) * s ^*(mod (i, 16)).

The spreading factor that establishing is put is SF _dpch, the symbolic number of data block 1 and data block 2 is: N _sym=352/SF _dpch;

According to configuration, the spreading code (having contained spreading code coefficient according to protocol package) that obtains using is: c _dpch=[c _dpch(0), c _dpch(1) ..., c _dpch(SF _dpch-1)];

To d _des1despreading obtains: d _{1, soft}=[d _{1, soft}(0), d _{1, soft}(1) ..., d _{1, soft}(N _sym-1)],

Wherein $d_{1,\mathrm{soft}}\left(i\right)=\underset{j=0}{\overset{{\mathrm{SF}}_{\mathrm{dpch}}-1}{\mathrm{\Σ}}}{d}_{\mathrm{des}1}(i*{\mathrm{SF}}_{\mathrm{dpch}}+j)*c_{\mathrm{dpch}}^{*}\left(\mathrm{mod}(j,{\mathrm{SF}}_{\mathrm{dpch}})\right);$

To d _des2despreading obtains: d _{2, soft}=[d _{2, soft}(0), d _{2, soft}(1) ..., d _{2, soft}(N _sym-1)],

Wherein $d_{2,\mathrm{soft}}\left(i\right)=\underset{j=0}{\overset{{\mathrm{SF}}_{\mathrm{dpch}}-1}{\mathrm{\Σ}}}{d}_{\mathrm{des}2}(i*{\mathrm{SF}}_{\mathrm{dpch}}+j)*c_{\mathrm{dpch}}^{*}\left(\mathrm{mod}(j,{\mathrm{SF}}_{\mathrm{dpch}})\right).$

Different from WCDMA, the DPCH data d of descrambling, despreading _{1, soft}and d _{2, soft}be still qpsk modulation signal, can adopt and ask the mode of 4 powers to remove modulation intelligence, obtain: d _{1, pow4}=[d _{1, pow4}(0), d _{1, pow4}(1) ..., d _{1, pow4}(N _sym-1)], d _{2, pow4}=[d _{2, pow4}(0), d _{2, pow4}(1) ..., d _{2, pow2}(N _sym-1)],

Wherein $d_{k,\mathrm{pow}4}\left(i\right)=d_{k,\mathrm{soft}}^4\left(i\right),i=\mathrm{0,1},...,N_{\mathrm{sym}}-1,k=\mathrm{1,2}.$

Calculate the correlation of two data blocks: $R=\underset{i=1}{\overset{N_{\mathrm{sym}}}{\mathrm{\Σ}}}{d}_{1,\mathrm{sqr}}\left(i\right){d*}_{2,\mathrm{sqr}}\left(i\right);$

Calculate residual frequency deviation $f_{e2}=\frac{1}{2\mathrm{\π}{T}_s(N_{\mathrm{sym}}+N_{\mathrm{mid}})}\arg \left(R\right)/4,$ Wherein $T_s=\mathrm{SF}*T_c=\frac{\mathrm{SF}}{1.28\×{10}^6}$ Second, be the symbol period of DPCH channel, N _midfor the number under character rate corresponding to the time span of midamble code, be 144/SF.

Step S409, determine total frequency deviation: f _e=f _e1+ f _e2;

Step S410, the total frequency deviation of basis are carried out frequency calibration.

The embodiment of the present invention is also corresponding provides a kind of frequency calibration device, as shown in Figure 5, comprising:

Rising edge determining unit 501, the psk modulation signal sending for receiving terminal, and the rising edge position of definite signal;

Grouped element 502, for according to rising edge position data intercept, and divides into groups to the data that intercept according to over-sampling rate;

Thick frequency deviation estimating unit 503, for determining the grouping of variance minimum of amplitude, determines thick frequency deviation deviant according to the data in this grouping, and carries out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to amplitude;

Thin frequency deviation estimating unit 504, for determining time slot original position according to pilot tone or training sequence, intercepts the data of a time slot and determines thin frequency deviation deviant according to these data;

Frequency calibration unit 505, for according to thick frequency deviation deviant and thin frequency deviation deviant and, carry out frequency calibration.

In the time that described signal is WCDMA signal, rising edge determining unit 501 also for:

The psk modulation signal sending at receiving terminal, and before the rising edge position of definite signal, terminal is configured, make terminal only send DPCCH signal, do not send DPDCH signal.

Rising edge determining unit 501 specifically for:

Determine the performance number of psk modulation signal sampling point sliding window;

Determine the rising edge position of signal according to the performance number of the sampling point sliding window of each sampled point.

Grouped element 502 specifically for:

Determine data intercept be r '=[r ' (0), r ' (1) ..., r ' (N '-1)], N '=OSR*N _fOE, wherein r ' (i)=r (i+I _start+ N _Δ* OSR), OSR is over-sampling rate, N _Δfor the surplus setting in advance, N _fOE≤ N _cfor the symbolic number of estimating in order to frequency deviation, N _cit is the number of chips (being 2560 for WCDMA, is 864 for TD-SCDMA) of a time slot.

To the front N of the data that intercept _var* OSR data are divided into groups, and obtain OSR group r ' ₀, r ' ₁..., r ' _oSR-1, equal 1 sampling data in every group, wherein, N _varfor the number of chips of each grouping, and N _var≤ N _fOE.

Thick frequency deviation estimating unit 503 is determined the grouping of the variance minimum of amplitude, determines thick frequency deviation deviant according to the data in this grouping, specifically comprises:

The operation of M power is carried out in the grouping of the variance minimum to amplitude, and wherein, M is modulation index;

To the data padding after the M power operation obtaining, carry out FFT conversion, wherein, counting in setting range of FFT conversion, makes FFT resolution in the scope of thin frequency deviation estimation;

Data after FFT conversion are carried out to FFT shifting function, make zero-frequency in center;

Determine the peak of the absolute value of the data after FFT shifting function;

Obtain thick frequency deviation deviant according to the value of the difference of peak and center and FFT resolution and M.

In the time that signal is WCDMA signal, thick frequency deviation estimating unit 503 is carried out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to amplitude, specifically comprises:

Determine that the data after thick compensate of frequency deviation are r _comp=[r _comp(0), r _comp(1) ..., r _comp(N _comp-1)], wherein, $r_{\mathrm{comp}}\left(i\right)=r(I_{\mathrm{start}}+I_{\mathrm{opt}}+i*\mathrm{OSR})e^{-j2\mathrm{\πi}{f}_{\mathrm{el}}{T}_c},i=\mathrm{0,1},...,N_{\mathrm{comp}}-1;$ T _cfor chip period, N _comp=N _c+ N _sync, N _cbe the number of chips of 1 time slot, N _syncfor carrying out the data window length of synchronous search, f _e1for thick frequency deviation deviant, I _optfor the sampling deviation of this grouped data;

In the time that signal is WCDMA signal, thin frequency deviation estimating unit 504 is determined time slot original position according to pilot tone, specifically comprises:

According to the pilot frequency format of configuration, obtain original pilots bit:

d _pilot＝[d _pilot(0)，d _pilot(1)，…，d _pilot(N _pilot-1)]；

By d _pilotcarry out BPSK modulation and obtain modulation symbol d ' _pilot=[d ' _pilot(0), d ' _pilot(1) ..., d ' _pilot(N _pilot-1)], d ' wherein _pilot(i)=-2d _pilot(i)+1;

To d ' _pilotcarry out spread spectrum: $d_{\mathrm{spr},\mathrm{pilot}}=d_{\mathrm{pilot}}^{\′}\⊗c_{\mathrm{ch},\mathrm{256,0}}=[d_{\mathrm{pilot}}^{\′}\left(0\right)*c_{\mathrm{ch},\mathrm{256,0}},d_{\mathrm{pilot}}^{\′}\left(1\right)*c_{\mathrm{ch},\mathrm{256,0}},...,d_{\mathrm{pilot}}^{\′}(N_{\mathrm{pilot}}-1)*c_{\mathrm{ch},\mathrm{256,0}}],$ Wherein, represent direct product, c _{ch, 256,0}for spreading code.

By the scrambler s of this time slot ₀=[s ₀(0), s ₀(1) ..., s ₀(2559)] carry out scrambling and obtain reference pilot signal: d _{scr, pilot}=[d _{scr, pilot}(0), d _{scr, pilot}(1) ..., d _{scr, pilot}(256*N _pilot-1)], d wherein _{scr, pilot}(i)=(j*d _{spr, pilot}(i)) * s ₀(i);

Determine correlation R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)], wherein,

$R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{N_{c,\mathrm{pilot}}-1}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j)d_{\mathrm{scr},\mathrm{pilot}}*\left(j\right);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is r (I _start+ I _opt+ I _max* OSR);

In the time that signal is TD-SCDMA signal, thin frequency deviation estimating unit 504 is determined time slot original position according to training sequence (intermediate code), specifically comprises:

Obtain with reference to intermediate code symbol with intermediate code deviant according to the basic midamble code of configuration number:

d _mid＝[d _mid(0)，d _mid(1)，…，d _mid(N _mid-1)]，

Wherein, N _mid=144 is intermediate code symbolic number, d _mid(i) 1 ,-1, j, value in-j};

Determine correlation: R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)],

Wherein, $R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{127}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j+352+16)d_{\mathrm{mid}}*(j+16);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is: r (I _start+ I _opt+ I _max* OSR).

In the time that signal is WCDMA signal, thin frequency deviation estimating unit 504 is determined thin frequency deviation deviant according to these data, specifically comprises:

These data are carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence;

In the time that described signal is TD-SCDMA signal, thin frequency deviation estimating unit 504 is determined thin frequency deviation deviant according to these data, specifically comprises:

Get the result after the thick compensate of frequency deviation of a time slot;

Intercept the data block of intermediate code both sides;

The data block intercepting is carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence.

The embodiment of the present invention provides a kind of transmitting frequency calibration method and device, make terminal send psk modulation signal, receive after psk modulation signal, intercept sufficiently long one piece of data behind rising edge position, one group of data determining the variance minimum of amplitude carry out that thick frequency deviation is estimated and thick compensate of frequency deviation, determine again time slot original position more accurately, again intercept the data of a time slot, determine thin frequency deviation deviant according to these data, thereby obtain frequency deviation skew more accurately, carry out frequency calibration according to this frequency deviation skew and can improve the accuracy of frequency calibration, simultaneously, owing to first carrying out, thick frequency deviation is estimated and thick compensate of frequency deviation, determine again time slot original position more accurately, data intercept carries out thin compensate of frequency deviation, so improved the efficiency of definite frequency deviation deviant, and then improve the efficiency of frequency calibration.

Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt complete hardware implementation example, completely implement software example or the form in conjunction with the embodiment of software and hardware aspect.And the present invention can adopt the form at one or more upper computer programs of implementing of computer-usable storage medium (including but not limited to magnetic disc store, CD-ROM, optical memory etc.) that wherein include computer usable program code.

The present invention is with reference to describing according to flow chart and/or the block diagram of the method for the embodiment of the present invention, equipment (system) and computer program.Should understand can be by the flow process in each flow process in computer program instructions realization flow figure and/or block diagram and/or square frame and flow chart and/or block diagram and/or the combination of square frame.Can provide these computer program instructions to the processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device to produce a machine, the instruction that makes to carry out by the processor of computer or other programmable data processing device produces the device for realizing the function of specifying at flow process of flow chart or multiple flow process and/or square frame of block diagram or multiple square frame.

These computer program instructions also can be stored in energy vectoring computer or the computer-readable memory of other programmable data processing device with ad hoc fashion work, the instruction that makes to be stored in this computer-readable memory produces the manufacture that comprises command device, and this command device is realized the function of specifying in flow process of flow chart or multiple flow process and/or square frame of block diagram or multiple square frame.

These computer program instructions also can be loaded in computer or other programmable data processing device, make to carry out sequence of operations step to produce computer implemented processing on computer or other programmable devices, thereby the instruction of carrying out is provided for realizing the step of the function of specifying in flow process of flow chart or multiple flow process and/or square frame of block diagram or multiple square frame on computer or other programmable devices.

Although described the preferred embodiments of the present invention, once those skilled in the art obtain the basic creative concept of cicada, can make other change and amendment to these embodiment.So claims are intended to be interpreted as comprising preferred embodiment and fall into all changes and the amendment of the scope of the invention.

Obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if these amendments of the present invention and within modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims (16)

1. a transmitting frequency calibration method, is characterized in that, comprising:

The psk modulation signal that receiving terminal sends, and the rising edge position of definite described signal;

According to rising edge position data intercept, and according to over-sampling rate, the data that intercept are divided into groups;

Determine the grouping of the variance minimum of amplitude, determine thick frequency deviation deviant according to the data in this grouping, and carry out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to described amplitude;

Determine time slot original position according to pilot tone or training sequence, intercept the data of a time slot and determine thin frequency deviation deviant according to these data;

According to described thick frequency deviation deviant and described thin frequency deviation deviant and, carry out frequency calibration.

2. the method for claim 1, is characterized in that, in the time that described signal is WCDMA signal, and the psk modulation signal that described receiving terminal sends, and determine before the rising edge position of described signal, also comprise:

Terminal is configured, makes terminal only send DPCCH signal, do not send DPDCH signal.

3. the method for claim 1, is characterized in that, the psk modulation signal that described receiving terminal sends, and the rising edge position of definite described signal, specifically comprise:

Determine the performance number of described psk modulation signal sampling point sliding window;

Determine the rising edge position I of described signal according to the performance number of the sampling point sliding window of each sampled point _start.

4. the method for claim 1, is characterized in that, described according to rising edge position data intercept, and according to over-sampling rate, the data that intercept is divided into groups, and specifically comprises:

Determine data intercept be r '=[r ' (0), r ' (1) ..., r ' (N '-1)], N '=OSR*N _fOE, wherein r ' (i)=r (i+I _start+ N _Δ* OSR), OSR is over-sampling rate, N _Δfor the surplus setting in advance, N _fOE≤ N _cfor the symbolic number of estimating in order to frequency deviation, N _cit is the number of chips of a time slot;

To the front N of the data that intercept _var* OSR data are divided into groups, and obtain OSR group r ' ₀, r ' ₁..., r ' _oSR-1, equal 1 sampling data in every group, wherein, N _varfor the number of chips of each grouping, and N _var≤ N _fOE.

5. method as claimed in claim 4, is characterized in that, the grouping of the variance minimum of described definite amplitude is determined thick frequency deviation deviant according to the data in this grouping, specifically comprises:

The operation of M power is carried out in the grouping of the variance minimum to amplitude, and wherein, M is modulation index;

To the data padding after the M power operation obtaining, carry out FFT conversion, wherein, counting in setting range of FFT conversion, makes FFT resolution in the scope of thin frequency deviation estimation;

Data after FFT conversion are carried out to FFT shifting function, make zero-frequency in center;

Determine the peak of the absolute value of the data after FFT shifting function;

Obtain thick frequency deviation deviant according to the value of the difference of peak and center and FFT resolution and M.

6. method as claimed in claim 5, is characterized in that, compensate of frequency deviation is carried out in the described grouping by the variance minimum of this thick frequency deviation deviant to described amplitude, specifically comprises:

Determine that the data after thick compensate of frequency deviation are r _comp=[r _comp(0), r _{comp (}1) ..., r _comp(N _comp-1)], wherein, $r_{\mathrm{comp}}\left(i\right)=r(I_{\mathrm{start}}+I_{\mathrm{opt}}+i*\mathrm{OSR})e^{-j2\mathrm{\πi}{f}_{\mathrm{el}}{T}_c},i=\mathrm{0,1},...,N_{\mathrm{comp}}-1;$ T _cfor chip period, N _comp=N _c+ N _sync, N _cbe the number of chips of 1 time slot, N _syncfor carrying out the data window length of synchronous search, f _e1for thick frequency deviation deviant, I _optfor the sampling deviation of this grouped data.

7. method as claimed in claim 6, is characterized in that, in the time that described signal is WCDMA signal, describedly determines time slot original position according to pilot tone, specifically comprises:

According to the pilot frequency format of configuration, obtain original pilots bit:

d _pilot＝[ _dpilot(0)，d _pilot(1)，…，d _pilot(N _pilot-1)]；

By d _pilotcarry out BPSK modulation and obtain modulation symbol d ' _pilot=[d ' _pilot(0), d ' _pilot(1) ..., d ' _pilot(N _pilot-1)], d ' wherein _pilot(i)=-2d _pilot(i)+1;

To d ' _pilotcarry out spread spectrum: $d_{\mathrm{spr},\mathrm{pilot}}=d_{\mathrm{pilot}}^{\′}\⊗c_{\mathrm{ch},\mathrm{256,0}}=[d_{\mathrm{pilot}}^{\′}\left(0\right)*c_{\mathrm{ch},\mathrm{256,0}},d_{\mathrm{pilot}}^{\′}\left(1\right)*c_{\mathrm{ch},\mathrm{256,0}},...,d_{\mathrm{pilot}}^{\′}(N_{\mathrm{pilot}}-1)*c_{\mathrm{ch},\mathrm{256,0}}],$ Wherein, represent direct product, c _{ch, 256,0}for spreading code;

By the scrambler s of this time slot ₀=[s ₀(0), s ₀(1) ..., s ₀(2559)] carry out scrambling and obtain reference pilot signal: d _{scr, pilot}=[d _{scr, pilot}(0), d _{scr, pilot}(1) ..., d _{scr, pilot}(256*N _pilot-1)], d wherein _{scr, pilot}(i)=(j*d _{spr, pilot}(i)) * s ₀(i);

Determine correlation: R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)], wherein,

$R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{N_{c,\mathrm{pilot}}-1}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j)d_{\mathrm{scr},\mathrm{pilot}}*\left(j\right);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is: r (I _start+ I _opt+ I _max* OSR);

In the time that described signal is TD-SCDMA signal, describedly determine time slot original position according to training sequence, specifically comprise:

Obtain with reference to intermediate code symbol with intermediate code deviant according to the basic midamble code of configuration number:

d _mid＝[d _mid(0)，d _mid(1)，…，d _mid(N _mid-1)]；

Wherein, N _mis=144 is intermediate code symbolic number, d _mid(i) 1 ,-1, j, value in-j};

Determine correlation: R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)],

Wherein, $R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{127}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j+352+16)d_{\mathrm{mid}}*(i+16);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is: r (I _start+ I _opt+ I _max* OSR).

8. method as claimed in claim 7, is characterized in that, in the time that described signal is WCDMA signal, describedly determines thin frequency deviation deviant according to these data, specifically comprises:

These data are carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence;

In the time that described signal is TD-SCDMA signal, describedly determine thin frequency deviation deviant according to these data, specifically comprise:

Get the result after the thick compensate of frequency deviation of a time slot;

Intercept the data block of intermediate code both sides;

The data block intercepting is carried out to descrambling, despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence.

9. a frequency calibration device, is characterized in that, comprising:

Rising edge determining unit, the psk modulation signal sending for receiving terminal, and the rising edge position of definite described signal;

Grouped element, for according to rising edge position data intercept, and divides into groups to the data that intercept according to over-sampling rate;

Thick frequency deviation estimating unit, for determining the grouping of variance minimum of amplitude, determines thick frequency deviation deviant according to the data in this grouping, and carries out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to described amplitude;

Thin frequency deviation estimating unit, for determining time slot original position according to pilot tone or training sequence, intercepts the data of a time slot and determines thin frequency deviation deviant according to these data;

Frequency calibration unit, for according to described thick frequency deviation deviant and described thin frequency deviation deviant and, carry out frequency calibration.

10. device as claimed in claim 9, is characterized in that, in the time that described signal is WCDMA signal, described rising edge determining unit also for:

The psk modulation signal sending at receiving terminal, and determine before the rising edge position of described signal, terminal is configured, make terminal only send DPCCH signal, do not send DPDCH signal.

11. devices as claimed in claim 9, is characterized in that, described rising edge determining unit specifically for:

Determine the performance number of described psk modulation signal sampling point sliding window;

Determine the rising edge position of described signal according to the performance number of the sampling point sliding window of each sampled point.

12. devices as claimed in claim 9, is characterized in that, described grouped element specifically for:

Determine data intercept be r '=[r ' (0), r, (1) ..., r ' (N '-1)], N '=OSR*N _fOE, wherein r ' (i)=r (i+I _start+ N _Δ* OSR), OSR is over-sampling rate, N _Δfor the surplus setting in advance, N _fOE≤ N _cfor the symbolic number of estimating in order to frequency deviation, N _cit is the number of chips of a time slot;

To the front N of the data that intercept _var* OSR data are divided into groups, and obtain OSR group r ' ₀, r ' ₁..., r ' _oSR-1, equal 1 sampling data in every group, wherein, N _varfor the number of chips of each grouping, and N _var≤ N _fOE.

13. devices as claimed in claim 12, is characterized in that, described thick frequency deviation estimating unit is determined the grouping of the variance minimum of amplitude, determine thick frequency deviation deviant according to the data in this grouping, specifically comprise:

The operation of M power is carried out in the grouping of the variance minimum to amplitude, and wherein, M is modulation index;

To the data padding after the M power operation obtaining, carry out FFT conversion, wherein, counting in setting range of FFT conversion, makes FFT resolution in the scope of thin frequency deviation estimation;

Data after FFT conversion are carried out to FFT shifting function, make zero-frequency in center;

Determine the peak of the absolute value of the data after FFT shifting function;

Obtain thick frequency deviation deviant according to the value of the difference of peak and center and FFT resolution and M.

14. devices as claimed in claim 13, is characterized in that, described thick frequency deviation estimating unit is carried out compensate of frequency deviation by the grouping of the variance minimum of this thick frequency deviation deviant to described amplitude, specifically comprises:

Determine that the data after thick compensate of frequency deviation are r _comp=[r _comp(0), r _comp(1) ..., r _{comp (}n _comp-1)], wherein, $r_{\mathrm{comp}}\left(i\right)=r(I_{\mathrm{start}}+I_{\mathrm{opt}}+i*\mathrm{OSR})e^{-j2\mathrm{\πi}{f}_{\mathrm{el}}{T}_c},i=\mathrm{0,1},...,N_{\mathrm{comp}}-1;$ T _cfor chip period, N _comp=N _c+ N _sync, N _cbe the number of chips of 1 time slot, N _syncfor carrying out the data window length of synchronous search, f _e1for thick frequency deviation deviant, I _optfor the sampling deviation of this grouped data.

15. devices as claimed in claim 14, is characterized in that, in the time that described signal is WCDMA signal, described thin frequency deviation estimating unit is determined time slot original position according to pilot tone, specifically comprises:

According to the pilot frequency format of configuration, obtain original pilots bit:

d _pilot＝[d _pilot(0)，d _pilot(1)，…，d _pilot(N _pilot-1)]；

By d _pilotcarry out BPSK modulation and obtain modulation symbol d ' _pilot=[d ' _pilot(0), d ' _pilot(1) ..., d ' _pilot(N _pilot-1)],

Wherein d ' _pilot(i)=-2d _pilot(i)+1;

To d ' _pilotcarry out spread spectrum: $d_{\mathrm{spr},\mathrm{pilot}}=d_{\mathrm{pilot}}^{\′}\⊗c_{\mathrm{ch},\mathrm{256,0}}=[d_{\mathrm{pilot}}^{\′}\left(0\right)*c_{\mathrm{ch},\mathrm{256,0}},d_{\mathrm{pilot}}^{\′}\left(1\right)*c_{\mathrm{ch},\mathrm{256,0}},...,d_{\mathrm{pilot}}^{\′}(N_{\mathrm{pilot}}-1)*c_{\mathrm{ch},\mathrm{256,0}}],$ Wherein, represent direct product, c _{ch, 256,0}for spreading code;

By the scrambler s of this time slot ₀=[s ₀(0), s ₀(1) ..., s ₀(2559)] carry out scrambling and obtain reference pilot signal: d _{scr, pilot}=[d _{scr, pilot}(0), d _{scr, pilot}(1) ..., d _{scr, pilot}(256*N _pilot-1)], d wherein _{scr, pilot}(i)=(j*d _{sr, pilot}(i)) * s ₀(i);

Determine correlation R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)], wherein,

$R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{N_{c,\mathrm{pilot}}-1}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j)d_{\mathrm{scr},\mathrm{pilot}}*\left(j\right);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is r (I _start+ I _opt+ I _max* OSR);

In the time that described signal is TD-SCDMA signal, described thin frequency deviation estimating unit is determined time slot original position according to training sequence, specifically comprises:

Obtain with reference to intermediate code symbol with intermediate code deviant according to the basic midamble code of configuration number:

d _mid＝[d _mid(0)，d _mid(1)，…，d _mid(N _mid-1)]，

Wherein, N _mid=144 is intermediate code symbolic number, d _mid(i) 1 ,-1, j, value in-j};

Determine correlation: R _pilot=[R _pilot(0), R _pilot(1) ..., R _pilot(N _sync-1)],

Wherein, $R_{\mathrm{pilot}}\left(i\right)=\underset{j=0}{\overset{127}{\mathrm{\Σ}}}{r}_{\mathrm{comp}}(i+j+352+16)d_{\mathrm{mid}}*(i+16);$

The peak of determining correlation is: $I_{\max }=\arg \underset{i=\mathrm{0,1},...,N_{\mathrm{sync}}-1}{\max }\left({\left|R_{\mathrm{pilot}}\left(i\right)\right|}^2\right);$

The accurate original position of determining time slot is: r (I _start+ I _opt+ I _max* OSR).

16. devices as claimed in claim 15, is characterized in that, in the time that described signal is WCDMA signal, described thin frequency deviation estimating unit is determined thin frequency deviation deviant according to these data, specifically comprises:

These data are carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence;

In the time that described signal is TD-SCDMA signal, described thin frequency deviation estimating unit is determined thin frequency deviation deviant according to these data, specifically comprises:

Get the result after the thick compensate of frequency deviation of a time slot;

Intercept the data block of intermediate code both sides;

The data block intercepting is carried out to descrambling and despreading;

Remove the modulation intelligence in the data after despreading;

Determine thin frequency deviation deviant according to the data of removing after modulation intelligence.